Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Riesz Neural Operator for Solving Partial Differential Equations
shouyiliu, Xiaokang Yang, Yuntian Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the local module to the global module (ablation) | RNO · Duffing | relative l2 error: 0.2098 -> 0.1801 | Improved |
| add the global module to the local module (ablation) | RNO · Duffing | relative l2 error: 0.2262 -> 0.1801 | Improved |
| add the mixer to global+local (ablation) | RNO · Duffing | relative l2 error: 0.1801 -> 0.1663 | Improved |
| add the mixer to global+local (ablation) | RNO · Reaction-Diffusion | relative l2 error: 0.0903 -> 0.0899 | Improved |
| use two orthogonal directions instead of one or three/five (ablation) | RNO · Beam, Reaction-Diffusion, Brusselator | relative l2 error | Improved |
| use more non-orthogonal directions (3 or 5) (ablation) | RNO · Beam, Reaction-Diffusion, Brusselator | relative l2 error | No clear effect |
| use ReLU instead of sin activation (ablation) | RNO · Duffing | relative l2 error: 0.1663 -> 0.1441 | Improved |
| remove the activation function (ablation) | RNO · Duffing | relative l2 error: 0.1663 -> 0.3215 | Hurt |
| remove the activation function (ablation) | LNO · Duffing | relative l2 error: 0.3325 -> 0.8121 | Hurt |
| use RNO instead of LNO (ablation) | Navier-Stokes Re = 5000 · Navier-Stokes | mean-squared error: 2.3139 -> 0.9121 | Improved |
| use RNO instead of LNO (ablation) | ERA5 · ERA5 | relative l2 error: 0.0062 -> 0.0022 | Improved |
| use RNO instead of second-best | Beam · Beam | relative l2 error: 0.0452 -> 0.0219 | Improved |
Closest papers by meaning
- Reference Neural Operators: Learning the Smooth Dependence of Solutions of PDEs on Geometric Deformations (ICML 2024 · similarity 0.87)
- Neural Operators with Localized Integral and Differential Kernels (ICML 2024 · similarity 0.87)
- Component Fourier Neural Operator for Singularly Perturbed Differential Equations (AAAI 2024 · similarity 0.87)
- Latent Mamba Operator for Partial Differential Equations (ICML 2025 · similarity 0.87)
- Neural Spectral Methods: Self-supervised learning in the spectral domain (ICLR 2024 · similarity 0.87)
- Derivative-enhanced Deep Operator Network (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
